Cardiac arrhythmias can be detected and treated by implantable cardioverter defibrillators (ICDs). ICDs typically monitor an intracardiac electrogram (EGM) signal to determine a patient's heart rhythm. When tachycardia or fibrillation are detected, electrical stimulation therapies are delivered, which may include pacing therapies and/or cardioversion/defibrillation shock therapies. Occasionally, a shock therapy is not successful in converting the arrhythmia back to normal sinus rhythm. In this case, another shock therapy is typically delivered as soon as possible. In order to determine success of a shock therapy, the ICD must detect the presence of the normal sinus rhythm. However, the EGM signal can be disrupted due to the shock delivery and other factors can delay detection of the heart rhythm and may result in unnecessarily repeating a shock therapy. The delivery of a shock therapy can be painful to the patient and uses considerable battery charge. As such, it is desirable to avoid continuing deliver of shock therapy when, for example, a preceding shock has been successful. Methods for evaluating the heart's response to a shock therapy that is not exclusively dependent on the EGM signal is needed.